Over the past few years, several devices for non-radiographic and non-invasive patient monitoring have been introduced such as, for example, Microsoft Kinect and Asus Xtion Pro. These devices, referred to herein as “RGB-D sensors.” offer a high degree of precision of 3D dense surface data in real time. The data captured from an RGB-D sensor can be utilized for assist medial image acquisition workflow in several ways. For example, using an RGB-D sensor in operating rooms, surgeons can navigate through medical images during procedures. In order to use the information obtained from the RGB-D sensor for medical scanner acquisition related tasks, the two imaging systems should be calibrated into a single coordinate system.
Any calibration approach must not impose any additional requirements on the environment where the medical scanner is placed. For instance, algorithms must not require specific size of the room or height of ceiling. This ensures that the techniques are applicable to both RGB-D sensors mounted on the medical scanners (possibly during assembly at industry) as well as sensors mounted on ceiling or external fixtures. In other words, the calibration should be carried out by technicians for each individual medical scanner during the scanner setup. Therefore, a simple calibration solution that does not require technicians to have knowledge of camera calibration is preferred.